Market leadership through technology - Kellogg School of ...

Market leadership through technology – Backward compatibility in the U.S. Handheld Video Game Industry

Jörg Claussen, Ifo Institute, Poschingerstr. 5, 81679 Munich, Germany, [email protected] Tobias Kretschmer, University of Munich and Ifo Institute, Schackstr. 4/III, 80539 Munich, Germany, [email protected] Thomas Spengler, University of Munich, Schackstr. 4/III, 80539 Munich, Germany, [email protected]

Keywords: backward compatibility, market leadership, network effects, video games, twosided markets. Acknowledgements: We thank participants at EARIE 2009, seminar participants at Universidad Carlos III Madrid, European Business School, EM Lyon, Hebrew University Jerusalem, INSEAD, LMU Munich, TU Munich, Tel Aviv University, and Albert BanalEstanol, Mélisande Cardona, Sofronis Clerides, Thorsten Grohsjean, Erik Lehmann, Tim Simcoe, Mariana Stamm, Jan Tonon, Stefan Wagner and Martin Watzinger for helpful comments and discussions. We acknowledge financial support from Deutsche Telekom Foundation and Vodafone R&D.

1

Market leadership through technology – Backward compatibility in the U.S. Handheld Video Game Industry

Abstract: The introduction of a new product generation forces incumbents in network industries to rebuild their installed base to maintain an advantage over potential entrants. We study if backward compatibility moderates this process of rebuilding an installed base. Using a structural model of the U.S. market for handheld game consoles, we show that backward compatibility lets incumbents transfer network effects from the old generation to the new to some extent but that it also reduces supply of new software. We examine the tradeoff between technological progress and backward compatibility and find that backward compatibility matters less if there is a large technological leap between two generations. We subsequently use our results to assess the role of backward compatibility as a strategy to sustain market leadership.

2

INTRODUCTION Firms frequently achieve a position of (temporary) market leadership in high-technology industries (Lee et al., 2010; Wade, 1995). Especially in the early stages of a technology’s life cycle, this can lead to a dominant design for a technology (Suarez and Utterback, 1995). In industries with significant network effects, the tendency for an industry (or a technological generation) to be dominated by a single firm or standard is especially strong (Arthur, 1989). This pattern of market leadership in technology-intensive industries creates incentives for firms to invest in continuous learning (Schilling, 2002), build a technological community (Wade, 1995), exploit complementarities across products (Lee et al., 2010), or engage in aggressive pricing (De Figueiredo and Silverman, 2007), among others. Securing an early lead can therefore translate into a competitive advantage throughout a technological generation (Arthur, 1989; Lieberman, 1989; Mascarenhas, 1992). Interestingly, in network industries, market leadership often persists across generations, which suggests that providers of successful technologies can carry over some of their dominance to future generations. The strategy literature identifies a number of mechanisms by which this intergenerational transfer of market leadership can take place. One stream of literature proposes that incumbent firms possess dynamic capabilities developed from prior generations that are useful in the new generation (Chen, Williams, and Agarwal, 2012; Danneels, 2002; De Figueiredo and Silverman, 2007; Eggers, 2012; Kotha, Zheng, and George, 2010). The core logic here is that firms are better at managing a new technology because they have learned how to do so in a prior one. Another line of research suggests that firms utilize complementary assets acquired in a previous product generation (Hill and Rothaermel, 2003; Jones, 2003; Rothaermel and Boeker, 2008; Tripsas, 1997). A specific form of leveraging complementary assets from a previous technological generation is to maintain backward compatibility with the old generation (Shapiro and Varian, 1999).

3

Our paper studies if backward compatibility by the market leader can be a strategy to sustain market leadership across generations. We pose three questions about the nature and implications of backward compatibility in markets with network effects: 1. How does backward compatibility influence demand for a new product generation? 2. How does backward compatibility affect the supply of new complementary products? 3. Are the effects of backward compatibility moderated by technological progress? We analyze the U.S. market for handheld game consoles, which is well-suited for our questions because i) backward compatibility is possible, but not necessary and ii) generation changes can be identified clearly. Handheld consoles are especially interesting as they exhibit different degrees of technological change across generations, so we can analyze the tradeoff between backward compatibility and technological progress across generations. Existing work in the market for home video game consoles studies asymmetric network effects (Shankar and Bayus, 2003), changes of indirect network effects over the product life cycle (Clements and Ohashi, 2005), software exclusivity (Corts and Lederman, 2009) and blockbuster software (Stremersch and Binken, 2009). While these papers handle multiple generations they do not explore how backward compatibility affects generational change and market leadership, except for Clements and Ohashi (2005), who address backward compatibility by adding available games for the Playstation 1 to those of the Playstation 2. The theoretical literature on cross-generational or ‘vertical’ compatibility (Katz and Shapiro, 1994) studies firm incentives for backward compatibility. Waldman (1993) and Choi (1994) find that price discrimination increases compatibility incentives, while Kende (1994) argues that backward compatibility is more likely as valuations for old and new technologies are similar and building an installed base of complementary products is expensive. These results are confirmed in a simulation model by Lee et al. (2003), who find that low valuation for backward compatibility and a small installed base advantage of the old generation render

4

backward compatibility less likely. Nahm (2008) finds that profits for the incumbent are higher with backward compatibility, which may increase its incentives to upgrade beyond the social optimum (Ellison and Fudenberg, 2000). From a demand perspective, Shy (1996) finds that backward compatibility increases the rate of new technology adoption. The empirical literature on cross-generational compatibility finds that backward compatibility helps carry over some installed base advantage to future generations. Liikanen et al. (2004) and Koski and Kretschmer (2005) study intergenerational effects between the first and second generations of mobile telephony and confirm the positive impact of backward compatibility. Greenstein (1993) finds that buyers are more likely to select a new mainframe computer system if they own a compatible predecessor system. Gandal et al. (2000) study audio technologies and run a counterfactual by assuming backward compatibility of CD and vinyl. Compatibility would have accelerated diffusion by 1.5 years. We estimate demand for handheld video consoles as well as supply of game titles. Our estimation strategy builds on Clements and Ohashi (2005), extending their approach to account for backward compatibility, console age and the level of technological progress from one generation to the next. Further, we identify console characteristics to allow for a meaningful comparison between the effects of backward compatibility and increased console performance. In line with prior work, we find that backward compatibility positively affects demand for a new generation. In addition, we find that: i) backward compatibility works through the installed base of software of the compatible parent generation, ii) it increases demand for hardware, but decreases supply of software, and iii) backward compatibility matters less if there is a large technological leap between two generations. We capture a (demand-enhancing) direct and a (demand-reducing) indirect effect of backward compatibility. The former directly influences the adoption decision through the installed base of software for the compatible parent generations. This effect weakens for

5

higher technological leaps between generations. The latter works indirectly as old software partly substitutes for new software and thus lowers new software demand, reducing software supply, which in turn decreases hardware demand. The demand-enhancing effect outweighs the demand-reducing effect so that backward compatibility helps transfer network effects across generations. Indeed, we find that the market leader, Nintendo, was able to maintain its market leadership across multiple generations through a strategy of backward compatibility. This paper is structured as follows. We first give an overview of the U.S. market for handheld game consoles. We then develop our hypotheses and test them using a model of hardware demand and software supply. A discussion of our results follows. Further, we analyze the role of backward compatibility in maintaining market leadership by performing a counterfactual experiment and considering alternative explanations. Finally, we conclude.

INDUSTRY BACKGROUND The market for handheld game consoles first took off with the appearance of Nintendo’s Game Boy in 1989, the first device to sell to the mass market (Forster, 2005). Handheld game consoles are – just as their (immobile) home video game counterparts – part of a system comprising both hard- and software. Hardware manufacturers supply consoles and often also software titles,1 while software providers concentrate on the development and distribution of games. Given indirect network effects (Clements and Ohashi, 2005), hardware suppliers have an interest to encourage development of complementary products, namely game titles. Since the ‘Atari shock’ in the early 1980s (when the game console market collapsed due to a sharp increase in poor game titles), hardware suppliers actively manage quality of the market’s software side: developers need to sign detailed licensing contracts which are then enforced by legal and technological means such as security chips (Genakos, 2001). This also prevents any

1

On average, hardware manufacturers produced 12.8% of game titles for their consoles. 6

hardware manufacturer from developing consoles that are compatible with games for other platforms. Our sample ranges from 1995 to 2007.2 Industry observers typically separate consoles into generations. In industry terminology, we study generations IV to VII (Forster, 2005). Table 1 provides an overview of the consoles in the generations we study. It is striking that Nintendo – from IV up to VII – was continuously present in the market while its competitors changed continuously. Figure 2 illustrates Nintendo’s market share dominance over the whole period. We now describe the competitive landscape over the four technology generations we cover. -----------------------------------------------------------INSERT TABLE 1 AND FIGURE 2 HERE -----------------------------------------------------------Generation IV comprised Nintendo’s Game Boy and Game Boy Pocket, and Sega’s Game Gear. At the start of our sample in 1995, these consoles had already been on the market for some time. The devices basically shared the market, with Nintendo’s share ranging between 60% and 80% and Sega’s moving between 20% and 40% accordingly. The generation V console Game Boy Pocket reached market shares exceeding 80% from 1998 on. This is remarkable considering that: i) the device was basically a remake with a smaller body but the same hardware capabilities as its predecessor, the Game Boy, and ii) Tiger Electronic’s Game.com, which had superior hardware capabilities, had also been launched in the meantime. Nintendo’s Virtual Boy – in contrast to the company’s other products – was comparably unsuccessful due to its bulkiness, problems during use3 and little

2

Extending the study period beyond 2007 would be problematic as smartphones (with Apple’s iPhone as the most prominent representative) have since then developed to be close substitutes to dedicated handheld game consoles. 3 Nintendo Virtual Boy’s image generation was based on a combination of a LED unit and oscillating mirrors. Users had to focus on these mirrors while playing which caused many 7

software available. It only reached substantial market share through a harsh price cut aimed at reducing stockpiles.4 The Game.com Pocket Pro, a lighter and less bulky remake of the Game.com, did not even reach 1% market share. The next dominant device was Nintendo’s Game Boy Color, which again was not the technically most advanced console of its time. Its main differentiating feature was the enormous installed base of backward compatible software titles from its predecessors. While its competitors did not have an installed base of existing games, the Game Boy Color could build on millions of software copies sold in the almost ten years the Game Boy platform had been on the market. Game Boy Color users did not have to wait for availability of new games and could buy or swap used games straight away. The next generation (VI) started with the Game Boy Advance. The device, which featured improved hardware power on the one hand and backward compatibility to Game Boy Color games on the other reached market shares close to 100% at the top of its cycle. While there was no device on the market at that time matching the Game Boy Advance in terms of hardware power, attributing its dominance merely to weak competition would be simplistic. Backward compatibility allowed users to draw on a game library comprising more than 46 million Game Boy Color titles from the outset, which clearly contributed to its success. In early 2003 Nintendo launched the Game Boy Advance SP, a facelifted Game Boy Advance with identical technology but a new body design and minor screen improvements. It matched its predecessor’s success, completely dominating the market at the top of its cycle. It prevailed not only over dated devices like the Neo Geo Pocket Color but also over Nokia’s NGage, which had a processor more than 6 times faster than the Game Boy Advance SP.

players headaches. Hence, the Virtual Boy bore a warning statement that it may cause headaches from the start of retail availability in the United States (Kent, 2002, pp. 513-515). 4 The maximum market share reached by the Virtual Boy was 44%, reached after cutting the initial price of more than $160 to less than $30 in April 1997. 8

At the end of 2004 Nintendo launched generation VII of handheld game consoles. Compared to the previous generation, the Nintendo DS was a significant improvement in terms of hardware performance. The device was again backward compatible and could play Nintendo’s generation VI games. However, in this generation Nintendo shared the market with Sony. Sony’s Playstation Portable (PSP) started with a market share exceeding 50% and then ranging between 20% and 40%. This is remarkable given that Sony had to start from scratch in the business while Nintendo again had a strong installed base of games. The PSP was the most powerful handheld console ever and outperformed the DS by far – for example, it was nearly five times as fast as Nintendo’s DS. At the end of our study period both players Nintendo and Sony launched remakes of their consoles: the DS Lite and the Playstation Portable Slim. Both are lighter and possess a smaller body than their predecessors. Throughout the generations we study, Nintendo was successful, except with the Virtual Boy. At least part of its success may be due to the enormous installed bases of games leveraged by the company through backward compatibility. Sony’s success suggests that such dominance may be overcome by significant technological progress. While many companies failed in challenging Nintendo with consoles roughly on par, Sony’s Playstation Portable, which outperformed Nintendo’s DS by far, gained substantial market share quickly.5

HYPOTHESES We now derive hypotheses on the effect of backward compatibility on hardware demand and software supply. We first discuss how backward compatibility works directly and indirectly. We then argue why we expect the effect of backward compatibility to be less strong for larger technological leaps. To our knowledge, we are the first to empirically test both the positive and negative effects of backward compatibility and the effect of technological progress.

5

Note that in this industry, success is typically measured in terms of market share. 9

Direct influence of backward compatibility on hardware demand When an incumbent launches a technologically improved product generation, it usually faces competition from two directions: from the incumbent’s parent generation and from products offered by competing firms. The larger the incumbent’s installed base and the more fragmented the new generation, the more difficult it is to overcome this startup problem, causing excess inertia (Farrell and Saloner, 1985; Kretschmer, 2008) or technological lockout (Schilling, 2002). In markets with indirect network effects, firms face a chicken-and-egg problem: it is not enough to offer a new video console; consumers also expect to choose from a wide variety of games for it (Clements and Ohashi, 2005; Corts and Lederman, 2009; Gandal et al., 2000; Gupta, Jain, and Sawhney, 1999). Gandal et al. (2000) identify three strategies for markets with indirect network effects to overcome startup problems. Firms can (1) subsidize hardware, (2) increase software availability by forward integration, and (3) make the product backward compatible with the parent generation. All three strategies are used in the videogame market. Especially shortly after product launch, consoles are often sold at or below marginal costs. Most console manufacturers also develop and publish games on their own to increase availability of software for their own consoles (Corts and Lederman, 2009). The strategy we focus on in this paper is the use of backward compatibility to transfer network effects across generations, also widely used in the video games industry. In the market for handheld game consoles, backward compatibility implies that game cartridges of the parent generation can still be used with the new console generation. If the physical format of the game cartridges changes, this may even require a second cartridge

10

slot.6 Backward compatibility is costly for the console manufacturer: the enclosure has to be bigger, additional parts are needed, and the processor must be able to process the old games. How will backward compatibility work exactly? Indirect network effects in the videogame industry have so far been measured through the demand-increasing effect by the number of games currently offered on the market (Clements and Ohashi, 2005; Corts and Lederman, 2009). One way to assess the effect of backward compatibility could be to analyze if and how much the number of games still available for the compatible parent generation affect demand for the new generation. Alternatively, the stock (or installed base) of games sold for the parent generation would proxy all games that could potentially be used with the new console. This measure, which we feel is more plausible since buying old games for a new console generation may be somewhat unattractive, implies that a larger installed base of compatible games increases the likelihood that a potential adopter has access to some of these games and thus benefit from backward compatibility. A person has access to old games if she owns the parent console or gets old games from friends or through second-hand trading.7 The results by Greenstein’s (1993) support our intuition in another setting: he finds the likelihood of adopting a new-generation mainframe to be higher if a firm owns a previous-generation mainframe of the same brand. Given the switch of mainframes is a binary (hardware) decision, while the software used on the previous mainframe can be used with the new mainframe, the availability of existing software, and not sales or software development post-hardware purchase, is what affects demand for hardware. This is summarized in our first hypothesis: Hypothesis 1:

Backward compatibility increases hardware demand more the higher the prior generation’s installed base of software.

6

This was the case for the Game Boy Advance, which had one slot for old Game Boy Color cartridges and one for new Game Boy Advance ones. 7 There is a sizable second-hand market for console games. E.g., on eBay.com, as of 24 January 2012, a total of 108,466 used games for mobile devices were offered. 11

The dark side of backward compatibility In addition to the direct effect of complementary goods on the baseline product, backward compatibility can also have a negative impact on the new generation. This negative impact stems from the fact that complementary goods from both generations are substitutes for each other. In the console market this implies that a new console can be used to run games from the current generation as well as compatible games from the previous one. Given that most games are provided by independent developers,8 this implies that developers of new games will face more competition – not only from competitors in the same generation, but also from their predecessors (Kretschmer, 2008). Given the fixed-cost nature of game development, developers will expect less revenues to cover their (sunk) fixed costs, so that their incentive to develop and release new games – basically, to enter the new generation – decreases (Sutton, 1998). Therefore, what may be beneficial for the hardware market because there is a stock of complementary goods available may be harmful for the software market because it decreases the incentives to develop software for the new generation. This is summarized in our second hypothesis: Hypothesis 2:

Backward compatibility decreases supply of software titles for the new generation more the higher the prior generation’s installed base of software.

This hypothesized effect represents the ‘dark side of compatibility’. Therefore, the net effect of backward compatibility is determined by two countervailing effects: First, the direct effect of backward compatibility suggests that availability of games for the compatible parent generation serves as a (part-)substitute for variety of new games, increasing hardware demand. Second, the indirect effect of backward compatibility implies that the substitution of

8

In our sample, only 12% of games were published by one of the hardware manufacturers. 12

new games by old games reduces new software demand, which in turn lowers software supply, which eventually reduces hardware demand. Backward compatibility and technological progress Our final set of hypotheses addresses the potential tradeoff between backward compatibility and technological progress. Shapiro and Varian (1999) identify this as the tradeoff between ‘evolution’ (ensuring backward compatibility but offering limited technological improvement) and ‘revolution’ (sacrificing backward compatibility, but offering drastically increased performance) strategies. However, Shapiro and Varian (1999) conceptualize these as decisions based on technological restrictions. The argument is that significant performance increases can only be secured by using the latest technology, which in turn makes it more difficult or costly to maintain backward compatibility. In our empirical setting, backward compatibility is achieved without any performance losses (i.e. an old game runs just as well on a new console as on an old one), so that one might assume that backward compatibility is perfect and that technological improvements affect demand in general, but not the effect of backward compatibility. In other words, for perfect compatibility the degree of substitutability between old and new-generation games is independent of the technological gap between the old and the new generation console. However, the degree of substitutability rests on the extent to which games exploit the technical capabilities of a particular console. As games for the old generation were designed with a different set of technological restrictions, new games will differ significantly in their performance especially if the set of restrictions imposed by the current generation console has changed considerably (Shy, 1996). Our empirical setting lets us identify the relative importance of technological improvement and backward compatibility if both are present. We expect technological progress and backward compatibility to be substitutes – however, unlike Shapiro and Varian (1999) we assume substitutability to be consumer-driven

13

rather technologically determined. The degree of substitutability of old and new games depends on the relative performance of the two game generations and backward compatibility. As a large technological improvement on the hardware side permits the design of better (i.e. more elaborately programmed) games, an old game will be a worse substitute as the technological frontier is pushed out. We thus expect technological progress to have a moderating effect on both the demand- and supply side effects of backward compatibility, which is summarized in Hypotheses 3a and 3b: Hypothesis 3a:

The positive effect of backward compatibility on hardware demand is negatively moderated by the degree of technological progress between generations.

Hypothesis 3b:

The negative effect of backward compatibility on software supply is negatively moderated by the degree of technological progress between generations.

DATA AND ESTIMATION MODEL Data Data sources The core data set for our analysis comes from the market research firm NPD Group and consists of monthly unit sales and revenues in the market for handheld game consoles in the U.S. for the period from 1/1995 to 11/2007.9 While, to the best of our knowledge, we are the first to use the data about handheld game consoles, NPD data on video consoles has already

9

We include hardware-only sales, i.e. just the console, and packages comprising a console and a game. Both are treated equally in the analysis as (i) package prices do not differ significantly from that of single consoles and (ii) a clear separation is not possible with our data. Moreover, many consoles are rarely sold on their own. 14

been used for several other studies (Clements and Ohashi, 2005; Corts and Lederman, 2009; Mollick, 2012; Shankar and Bayus, 2003; Stremersch and Binken, 2009). Data on games for the different platforms is also supplied by NPD Group. The software data consists of monthly unit sales and revenue data for all available game titles. For each game title, the associated platform is reported. Note that game data is assigned on a platform (not console) level. We define a platform by a common game format. A platform can therefore consist of a single console (as for the Game Boy Color) or a family of consoles (as for the Game Boy and Game Boy Pocket) that use the same game format but are distinct regarding their hardware sales.10 Data on technical characteristics of the different consoles are also matched to our data. We use two variables representing the key dimensions that influence user perception: CPU speed as a proxy for processing power of the console and weight as a proxy for the console’s mobility. The major data source for these technical characteristics is Forster (2005, pp. 212-214). This is completed with specifications from suppliers’ websites, console databases and console information websites. All prices are deflated to enable comparison of console and game prices over the entire period. We use the U.S. deflator provided by the International Monetary Fund.11 We use monthly population estimates from the U.S. census bureau to proxy for market potential. Finally, we use USD-JPY exchange rates from the Pacific Exchange Rate Service12 for a price instrument discussed later.

10

The other platforms consisting of two consoles are Game Boy Advance and Game Boy Advance SP, Nintendo DS and Nintendo DS Lite, game.com and game.com Pocket Pro, NGage and N-Gage QD, as well as Playstation Portable and Playstation Portable Slim. There are no platforms with three or more consoles in our data set. 11 We used the International Monetary Fund’s World Economic Outlook Database for this. 12 Available at http://fx.sauder.ubc.ca/. 15

Variables The variables are described in Table 2 and Table 3 reports summary statistics. In line with Corts and Lederman (2009), we eliminate the influence from outdated consoles selling remainders or products that never reached a wider audience by considering only devices that sold more than 500 units in a given month.13 --------------------------------------------------------------------INSERT TABLE 2 AND TABLE 3 HERE --------------------------------------------------------------------Market shares in the market for handheld game consoles are directly ( ) calculated by dividing the monthly unit sales of console by the total units sold in a given month. To derive

and

, we have to define potential market size first. Unlike Clements

and Ohashi (2005), who use the TV households to determine the number of potential buyers, we use the U.S. population numbers as several people in a household can own handheld consoles and handheld use is independent of TV ownership. From this, we derive is a console’s market share of the market potential14 and

, which

, the market share of the outside

good, i.e. the share of potential consumers that do not have a console and do not buy one in the given time period. By cumulating the unit sales data of hardware sales, we also derive each platform’s hardware installed base

15 16.

calculate each console’s average monthly price Software variety

Finally, we divide revenue by units to

. All prices are reported in 1995 USD.

is taken from the NPD data. For every platform we count the

number of game titles with positive sales to obtain 13

. Therefore,

can decline over time

The mean monthly total number of units sold is 627,068. The market potential is defined as the size of the population minus the number of people who already bought a handheld console. 15 We do not depreciate the installed base as (absolute) console performance does not deteriorate over time. 16 At the start of our dataset (1/1995), Nintendo’s Game Boy and Sega’s Game Gear have had already been on the market since 8/1998 and 1/1991. We therefore use data from http://vgchartz.com to derive the initial installed base of 12.7 respectively 2.9 million units for the Game Boy and the Game Gear. Data is derived by weighing the lifetime sales for Americas with the consoles’ 1995 U.S. share from total Americas sales. 14

16

if game titles are no longer sold. We also create the software installed base of the compatible preceding generation

.17

The last set of variables concerns the hardware characteristics of the handheld consoles. The dataset covers a twelve-year period in which technological progress for handheld game consoles was remarkable. For example, the mean CPU speed of active consoles grew from 3.93 MHz in 01/1995 to 187.43 MHz in 12/2007. As the data covers the entire period this causes problems in comparing devices’ capabilities. Comparing a 2007 console that is technically below average to the best device from 1995 would make the first one look far too good. We therefore normalize all variables containing technical data by the characteristics of contemporaneously active consoles. This is done by calculating yearly mean values and standard deviations for CPU speed and console weight. The yearly mean values and standard deviations obtained were then used to construct a z-score for each console. Finally,

is derived as the percentage improvement of the CPU speed

compared to the CPU speed of the compatible parent generation.18 Model specification We estimate both hardware demand and software supply. In line with prior work on indirect network effects, we use a structural model to estimate hardware demand and a reduced-form model to estimate software supply (Clements and Ohashi, 2005; Corts and Lederman, 2009; Nair, Chintagunta, and Dube, 2004). The two estimation models are derived below.

17

As for the hardware installed base, the software installed base for Game Boy and Game Gear is not directly available in our dataset. We therefore assume that the number of software titles sold per console in the years prior to the beginning of our dataset equals the number of software titles sold for each console in 1995. 18 We set this variable to zero if there is no active parent generation. 17

Hardware demand We model the demand side of the market using a structural model for hardware demand. Our model extends the discrete-choice model for differentiated products used by Clements and Ohashi (2005) and Corts and Lederman (2009) with measures of backward compatibility. We assume that each potential adopter of handheld video consoles maximizes its utility by choosing the highest

where

represents the different handheld consoles and

represents the outside option of not buying a console. The consumer’s utility function has the following (additive) functional form:

[ [

]

(1)

]

The first part of the utility function represents the baseline model that does not consider backward compatibility: utility depends on observed product characteristics console price

, software variety

19

unobserved characteristics

, the

, and the idiosyncratic

, which can be interpreted as the difference of consumer ’s valuation and the

error term mean utility.

This model is extended to capture the effects of backward compatibility. First, the installed base

of the prior generation’s compatible games is added. This variable is

used to test Hypothesis 1 and we expect it to have a positive influence on the buyer’s selection decision. Second, we add the improvement factor over the compatible parent and its interaction with installed base

. The improvement

factor expresses the relative increase in CPU speed compared to the CPU speed of the earlier

19

As already noted we distinguish between consoles and platforms multiple consoles using the same game format.

18

which can consist of

generation. We expect

to have a positive effect on utility as a technological leap

stimulates demand for a new product generation. In line with Hypothesis 3a however, we expect the interaction term to have a negative effect on the buyer’s utility. Further, we add console age

as well as an interaction term of installed base and console age,

.

For console age, we expect a negative influence as older consoles are less attractive to the remaining non-adopters. We also expect a negative coefficient for the interaction term between console age and installed base, as we expect the effect of a backward compatible installed base to be more important in the launch phase of a new platform generation. As in Clements and Ohashi (2005), we assume

to be identically and

independently distributed with an extreme value distribution function to generate a nested logit model (Berry, 1994). Potential adopters decide first to buy a handheld game console or not, and if they decide to buy one, they then select a specific console. In contrast to a simple logit model, substitution patterns can therefore differ between the decision of buying a console and the decision which console to buy. Setting the utility of the outside good equal to zero (Berry, 1994), we get a linear regression equation: (

)

(

)

( [ [

( )

) ]

(2)

]

Software supply We follow prior literature when estimating software supply (Clements and Ohashi, 2005; Corts and Lederman, 2009). Software supply is expressed by the variety of different game titles

available for a specific platform. We estimate the following reduced-form equation: [

]

19

(3)

[

]

[

]

The first line of the equation is the base model with

being brand-specific dummies,

the installed base of console of the current generation,

the age of the platform, and

an error term. We allow hardware installed base to interact with platform age (Clements and Ohashi, 2005). We extend the model with the same measures of backward compatibility as for the demand estimation. Following Hypothesis 2, we expect

to negatively affect

software supply as the installed base of backward compatible software might partly substitute for demand for new game titles. Further, from Hypotheses 3b we expect the interaction term of

with relative performance increase to be positive as they reduce the importance of

backward compatibility on the demand side and we therefore expect less substitution. Finally, we again allow for the interaction of our measure of backward compatibility with platform age to allow for changing importance of backward compatibility over time. Instruments Hardware demand The potential endogeneity of the three variables within-group share software variety

( )

, price

, and

requires the identification of appropriate instruments. We use the set of

instruments proposed by Clements and Ohashi (2005) and Corts and Lederman (2009). Within-group share is obviously correlated with the error term dependent variable

. As

as it contains part of the

is known to firms and consumers in the market (but not to the

econometrician), differences in unobserved quality might lead to different price setting and thus a correlation of the console price positive correlation between

and

. Finally, autocorrelation of

and the measure of software variety

leads to a

.

First, we use exchange rates between the U.S. and Japan as a cost side instrument for prices as many consoles come from Japan. Exchange rates seem a valid price instrument as 20

their change would probably lead to price adjustment in the U.S. market. However, it does not allow for identifying effects at the console level. Further, we use the average age of software titles currently available on the market to instrument for within-group share and console price. A high average age of games is a sign for missing supply of new game titles. Hence, we expect negative correlations of average software age both with within-group share as a lack of new games reduces the console’s relative attractiveness and with console price as console manufacturers may try to reduce counter this adverse effect by lowering prices. Finally, we construct several instruments that measure the extent of competition faced by a platform (Berry, Levinsohn, and Pakes, 1995). We use the sum of competing hardware characteristics,20 the total number of competing platforms, the number of competing platforms within a company, and the number of competing platforms within the same generation as instruments. Following Corts and Lederman (2009), these instruments are expected to be correlated with each of the three endogenous variables: with the within-group share as they affect utility of different options, with software variety as they influence incentives to provide game titles, and with price as they affect the ability to raise prices. Software supply The installed base of hardware

is possibly endogenous as unobserved shocks in the

software market might lead to increased software entry but also to increased hardware adoption. We use the instruments proposed by Clements and Ohashi (2005) to account for endogeneity. The average age of software titles on the market serves as an instrument, although the direction in which the instrument works is not clear. A high average software age could either indicate profitable opportunities or tough competition. We also use squared

20

We use the sums of the competing consoles’ cumulative CPU speed and weight. 21

platform age and an interaction term between platform age and average software age as supply-side instruments.

RESULTS The 2SLS estimation results are reported in Table 4 (hardware demand) and Table 5 (software supply).21 Columns 4-1 and 5-1 report results without the software installed base, 4-2 and 5-2 include just the linear term of the software installed base, and 4-3 and 5-3 include both the interaction terms and the hardware improvement factor. In all specifications, we use brand dummies to control for unobserved brand-specific effects as well as calendar month dummies to control for the strong seasonality in console sales. All 2SLS estimations are robust to arbitrary heteroskedasticity and arbitrary autocorrelation. -------------------------------------------------------------INSERT TABLE 4 AND TABLE 5 HERE -------------------------------------------------------------We first discuss results for the direct effect of backward compatibility (Hypothesis 1 and 2) before turning to the moderating effect of technological progress (Hypothesis 3a/3b). The control variables in the instrumented estimation results have the expected signs over the different specifications. With increasing age, demand as well as supply goes down. The interaction term of age and the installed base of compatible parent software is only significant for the demand side, confirming the expectation that the importance of backward compatibility declines over time. Further, the industry exhibits indirect network effects as the availability of more software variety

positively influences demand and the availability of

a larger hardware installed base in turn increases software variety. We find negative price elasticity of demand and a strong positive seasonal effect (not reported) in November and

21

The corresponding OLS regression results are available from the authors and show the same sign and significance for the hypothesis tests. 22

December for both demand and supply.22 Higher CPU speed increases demand, while higher console weight generally decreases demand. These results give us confidence in our model. Effect of backward compatibility on demand and supply We now discuss the first-order effect of backward compatibility on demand and supply. As discussed above, we expect backward compatibility to work through the installed base

of games for the parent generation.

Hardware demand We first observe that

has a significantly positive coefficient for both specifications (4-

2) and (4-3), which supports Hypothesis 1. For specification (4-3), we compare the effect of backward compatibility with indirect network effects from software variety

: one extra

game title for the current generation has the same impact on demand as 82,979 game titles sold for the parent generation.23 Applying this to the case of the Game Boy Advance, at the launch in June 2001 an installed base of 45.6 million compatible Game Boy Color games corresponded to the availability of 550 game titles for the new generation. In fact, at launch only 21 game titles were available for the Game Boy Advance and it took until August 2004 for 550 game titles to be available on the market. Software supply Adding

to the baseline specification as in estimation (5-2), we do not see any

significant effect from backward compatibility. However, in the full specification (5-3), we

22

As the right-hand side of the demand model is the mean utility of console in month , the magnitudes of the coefficients for the demand model cannot be interpreted in a meaningful way (Corts and Lederman, 2009). We therefore compare the strengths of different effects or discuss marginal effects from exogenous changes of a console’s backward compatibility. 23 The average unit sales of games in our sample are 118,619. 23

obtain a significant negative effect of

on software variety24. For each million games in

the installed base, 1.2 game titles less would be offered on the market. Again looking at the example of the Game Boy Advance, the installed base of 45.6 million compatible Game Boy Color games would reduce software supply by 54 titles at its launch date. This implies that absent an installed base, 75 games would have been available immediately from the launch of the Game Boy Advance. Backward compatibility and technological progress After discussing the direct effect of backward compatibility, we now turn to the interaction between backward compatibility and technological progress. Hardware demand Our results support Hypothesis 3a, as the interaction term has a significantly negative coefficient. Combining the counteracting effects of the installed base and the interaction term for specification (4-3), we see that backward compatibility has a positive effect if the percentage increase in CPU speed compared to the compatible parent generation is smaller than 363%. The largest technological leap between two succeeding generations in our data set is the switch from the Game Boy Advance SP to the Nintendo DS. For this generation change, CPU speed increased from 16.7 MHz to 67 MHz, an increase of 301%. Here, backward compatibility played a much weaker (although still positive) role. This coincides with the observation that the Playstation Portable, which entered the market four months later, was the only console to successfully challenge Nintendo’s dominance in handheld consoles market – with a much improved technology and up against a less influential installed base.

24

This is intuitive as we find a time-varying effect in (5-3), suggesting that a simple linear term is misspecified. Indeed, we find strong serial correlation in the error term in specification (5-1). 24

Software supply The results from specification (5-3) strongly support Hypothesis 3b that higher technological progress between generations reduces the supply-decreasing effect of backward compatibility. We see a substitutive effect from backward compatibility as long as the technological leap is smaller than 198%. Therefore, the Nintendo DS with an increase in CPU speed of 301% more than outweighs the substitutive effect.

BACKWARD COMPATIBILITY TO SUSTAIN DOMINANCE Our results suggest a strong effect of backward compatibility on the demand of new hardware generations. Since Nintendo was the only firm to launch successive console generations and therefore the only firm to report a positive installed base of backward compatible games, we ask if backward compatibility was a useful means of sustaining a dominant market position over multiple product generations. To isolate this effect however, we need to rule out that backward compatibility simply proxies for other unobserved factors – the Nintendo effect. We address this in two ways: First, we discuss the brand dummies in our regressions that aim to capture unobserved, brand-specific factors. Second, we run a counterfactual experiment by assigning one of the unsuccessful consoles, the Game.com console, the installed base of the then dominant console, the Game Boy. The Nintendo effect Table 6 reports the brand dummies for all players in the handheld game console market, with Nintendo the base category. ------------------------------------------------INSERT TABLE 6 HERE ------------------------------------------------While Sony’s brand dummy has a positive and significant sign – suggesting that both brand equity and technological advance played a role in successfully challenging Nintendo, the other dummies show no clear pattern. This implies that Nintendo’s reputation does not

25

significantly explain its success in repeatedly holding off competition. One explanation for Sony’s success (and the others’ failure) would be that Nintendo’s reputation suffered significantly just prior to the introduction of the PSP, which would lead to a significant and positive brand dummy for Sony as it measures the reputation relative to Nintendo. However, there is no anecdotal evidence for this in the relevant time period. Another consideration is that Nintendo’s reputation may have grown over time and that the backward compatible installed base (which grew more or less constantly throughout our sample) simply proxies for this reputation increase rather than a ‘real’ effect of backward compatibility. However, as the Sony PSP entered at the very end of the sample, this would make its success all the more improbable as it would have to be based on an implausibly high brand reputation vis-à-vis Nintendo. However, to alleviate this possible bias, we run our preferred regressions (4-3 and 5-3) using the rolling software installed base

of the

three years before the observation month instead of the overall installed base. The results are shown in Table 7 and show a qualitatively similar picture as our baseline results, rendering this alternative explanation less likely. ------------------------------------------------INSERT TABLE 7 HERE ------------------------------------------------A counterfactual experiment To assess if backward compatibility could indeed have played a role in sustaining Nintendo’s market leadership by intensifying the startup problem for challenging platforms, we run a counterfactual experiment in which we hypothetically assume that games for the Game Boy generation can be played on the Game.com console (and Nintendo consoles).25 In reality, the Game.com console was not backward compatible to any other parent console and was a

25

Such a move of mandating compatibility with a promising entrant could also be imposed by an antitrust authority as a pro-competitive measure (Shapiro, 1996). 26

commercial failure. Following Corts and Lederman (2009), we derive the counterfactual as follows. First, mean utility

for console at time is derived from the regression results of

our preferred specification (4-3). With the nested logit formula discussed in Berry (1994), the implied market shares can be obtained as follows: ( ⁄( ( with

∑

(

⁄(

)) )

(4)

)). In a next step, we assume that the Game.com

console, which was launched in 9/1997 could have played titles for the Game Boy. The installed base of compatible software titles for the parent generation increase of the Game.com CPU compared to the Game Boy CPU

, the performance , and the

interaction terms from equation (2) are adjusted accordingly. We then use the updated values to recalculate mean utilities and implied market shares. We repeat these steps for every month in the first year since the launch of the Game.com console and report average changes and the actual outcome in the top half of Table 8. ------------------------------------------------INSERT TABLE 8 HERE ------------------------------------------------First off, we observe that backward compatibility leads to an increase in total demand: the average additional demand of 231,707 Game.com units is twice as large as the aggregate decrease in demand for the competing platforms of 114,809 units. This is intuitive as overall network effects have increased. Without backward compatibility, the technologically superior Game.com never takes off and the outdated Game Boy Pocket maintains a dominant position, as can be seen from Game.com’s actual market share of 2.68%. Assigning the Game Boy’s installed base to Game.com changes the dynamics of the market drastically, and Game.com’s counterfactual market share is almost as high as Nintendo’s actual one.26

26

Note that we maintain Game Boy’s backward compatibility so that Game Boy and Game.com have equal installed bases. 27

In the bottom part of Table 8, we add the indirect effect of backward compatibility, which we found to decrease supply of new games. We proceed as follows. We first simulate backward compatibility of the Game.com console by changing the installed bases analogous to hardware demand. We then use the coefficients from our supply estimation (5-3) to predict the number of available games games in the utility function

. We finally substitute this (lower) number of available of the demand-side equation and can again derive implied

changes in units sold and in market shares. The indirect effect moderates the direct effect somewhat (as game providers for Game.com would have been deterred by the installed base of backward compatible games serving as imperfect substitutes). However, the direct effect dominates the indirect effect, so that backward compatibility would still have helped the Game.com console capture a large chunk of the market at the time.

CONCLUSION In this paper we study the effects of backward compatibility in a market with indirect network effects, the U.S. handheld game console industry. Backward compatibility helped the market leader Nintendo maintain their market leadership over multiple generations despite having an inferior technology in many instances. Backward compatibility in this market works through the installed base of games for a compatible parent generation and its strength is affected by the rate of technological improvement between successive generations. On the demand side, our results lend support to the role of backward compatibility. If a new generation is backward compatible with the old one, the installed base of games for the prior generation increases sales for the new generation console. However, large technological improvements across generations come at the cost of consumers valuing backward compatibility less as their utility from using the old complementary products is comparatively

28

low. Therefore, benefits from large technological improvement are partially offset by the reduced benefits from backward compatibility. On the supply side, we find that backward compatibility lowers the supply of new software, and that this effect is less pronounced for consoles with higher technological progress, which suggests that there is a (previously not identified) “dark side” to backward compatibility. By jointly analyzing hardware demand and software supply, we identify a tradeoff between the demand-enhancing effect of backward compatibility directly affecting hardware demand and the demand-reducing effect that works indirectly through reduced software variety for a platform. We find that the demand-increasing effect clearly outweighs the demand-decreasing effect. Could backward compatibility have stabilized market structure in the U.S. handheld console market by giving Nintendo a head start for every new generation? Sony’s PSP, the most successful challenger, entered with a much superior technology at a time when Nintendo had just made a significant technological leap from their previous generation, which is in line with our results that backward compatibility matters less if the generations are very different technologically, so that Nintendo was comparably more vulnerable at that junction. To further substantiate the claim that backward compatibility helped Nintendo maintain a dominant position over technologically superior challengers, we run a counterfactual and assign Nintendo’s Game Boy installed base to a technologically superior, but ultimately unsuccessful challenger, the Game.com console. We find that if Game.com had been backward compatible, market leadership would have been reversed. Our work relates to the literature on entry deterrence, as backward compatibility can discourage firms from entering a market or at least prevent them from attaining large market shares. While there are many theoretical models of strategic entry deterrence (Dixit, 1980; Haan, 2003; Klemperer, 1987; Milgrom and Roberts, 1982; Salop, 1979), empirical studies of

29

entry deterrence are rare in industrial organization (Schmalensee, 1978; Smiley, 1988). Strategy and marketing scholars have focused on limit pricing (Srinivasan, 1991), reputation (Clark and Montgomery, 1998) and excess capacity (Harrigan, 1981), while Gruca and Sudharshan (1995) integrate a wide variety of entry deterrence strategies in their conceptual framework, in part referring to product portfolio choices (brand proliferation, preannouncement, switching costs). However, technological parameters are not typically considered potential strategic instruments for entry deterrence.27 This is surprising as in technology-intensive industries entry is a salient phenomenon, often replacing current leaders in the process of creative destruction (Schumpeter, 1942). In this light, our results have both managerial and policy implications. Managers in network industries must consider backward compatibility an important parameter that helps stabilize market shares across generations and establish persistent market leadership. Judiciously managing the tradeoff between backward compatibility and technological progress is thus a key challenge for technology strategists.

27

An exception is Church and Gandal (1996), who study compatibility as a means of entry deterrence in a theoretical model. 30

REFERENCES Arthur W. 1989. Competing technologies, increasing returns and lock-in by historical events. Economic Journal 99(394): 116-131. Berry S. 1994. Estimating Discrete-Choice Models of Product Differentiation. RAND Journal of Economics 25(2): 242-262. Berry S, Levinsohn J, Pakes A. 1995. Automobile prices in market equilibrium. Econometrica 63(4): 841-890. Chen P-L, Williams C, Agarwal R. 2012. Growing pains: Pre-entry experience and the challenge of transition to incumbency. Strategic Management Journal 33(3): 252-276. Choi J. 1994. Network externality, compatibility choice, and planned obsolescence. Journal of Industrial Economics 42(2): 167-182. Church J, Gandal N. 1996. Strategic entry deterrence: Complementary products as installed base. European Journal of Political Economy 12(2): 331-354. Clark B, Montgomery D. 1998. Deterrence, reputations, and competitive cognition. Management Science 44(1): 62-82. Clements M, Ohashi H. 2005. Indirect network effects and the product cycle: video games in the U.S., 1994-2002. Journal of Industrial Economics 53(4): 515-542. Corts KS, Lederman M. 2009. Software exclusivity and the scope of indirect network effects in the U.S. home video game market. International Journal of Industrial Organization 27(2): 121-136. Danneels E. 2002. The dynamics of product innovation and firm competences. Strategic Management Journal 23(12): 1095-1121. De Figueiredo JM, Silverman BS. 2007. Churn, baby, churn: Strategic dynamics among dominant and fringe firms in a segmented industry. Management Science 53(4): 632-650. Dixit A. 1980. The role of investment in entry-deterrence. The Economic Journal 90(357): 95-106. Eggers JP. 2012. All experience is not created equal: learning, adapting, and focusing in product portfolio management. Strategic Management Journal 33(3). Ellison G, Fudenberg D. 2000. The neo-Luddite's lament: Excessive upgrades in the software industry. Rand Journal of Economics 31(2): 253-272. Farrell J, Saloner G. 1985. Standardization, compatibility, and innovation. Rand Journal of Economics 16(1): 70-83. Forster W. 2005. The Encyclopedia of game-machines: consoles, handhelds and home computers 1972-2005 (First Edition ed.). Gameplan.

31

Gandal N, Kende M, Rob R. 2000. The dynamics of technological adoption in hardware/software systems: The case of compact disc players. Rand Journal of Economics 31(1): 43-61. Genakos C. 2001. Game Wars - The Home Video Game Industry. In London Business School Case Study. Greenstein S. 1993. Did Installed Base Give an Incumbent any (Measureable) Advantages in Federal Computer Procurement? Rand Journal of Economics 24(1): 19-39. Gruca T, Sudharshan D. 1995. A framework for entry deterrence strategy: the competitive environment, choices, and consequences. Journal of Marketing 59(3): 44-55. Gupta S, Jain DC, Sawhney MS. 1999. Modeling the evolution of markets with indirect network externalities: An application to digital television. Marketing Science 18(3): 396-416. Haan M. 2003. Vaporware as a means of entry deterrence. Journal of Industrial Economics 51(3): 345-358. Harrigan K. 1981. Barriers to entry and competitive strategies. Strategic Management Journal 2(4): 395-412. Hill CWL, Rothaermel FT. 2003. The Performance of Incumbent Firms in the Face of Radical Technological Innovation. Academy of Management Review 28(2): 257-274. Jones N. 2003. Competing after radical technological change: the significance of product line management strategy. Strategic Management Journal 24(13): 1265-1287. Katz M, Shapiro C. 1994. Systems competition and network effects. Journal of Economic Perspectives 8(2): 93-115. Kende M. 1994. A note on backward compatibility. Economic Letters 45(3): 385-389. Kent SL. 2002. The ultimate history of video games: from Pong to Pokemon and beyond... Prima Publishing: Roseville (California). Klemperer P. 1987. Entry deterrence in markets with consumer switching costs. The Economic Journal 97(Conference 1987): 99-117. Koski H, Kretschmer T. 2005. Entry, standards and competition: Firm strategies and the diffusion of mobile telephony. Review of Industrial Organization 26(1): 89-113. Kotha R, Zheng Y, George G. 2010. Entry into new niches: the effects of firm age and the expansion of technological capabilities on innovative output and impact. Strategic Management Journal 32(9). Kretschmer T. 2008. Splintering and inertia in network industries. Journal of Industrial Economics 56(4): 685-706. Lee CH, Venkatraman N, Tanriverdi H, Iyer B. 2010. Complementarity-based hypercompetition in the software industry: Theory and empirical test, 1990–2002. Strategic Management Journal 31(13): 1431-1456.

32

Lee J, Lee J, Lee H. 2003. Exploration and Exploitation in the Presence of Network Externalities. Management Science 49(4): 553-570. Lieberman MB. 1989. The learning curve, technology barriers to entry, and competitive survival in the chemical processing industries. Strategic Management Journal 10(5): 431447. Liikanen J, Stoneman P, Toivanen O. 2004. Intergenerational effects in the diffusion of new technology: the case of mobile phones. International Journal of Industrial Organization 22(8-9): 1137-1154. Mascarenhas B. 1992. Order of entry and performance in international markets. Strategic Management Journal 13(7): 499-510. Milgrom P, Roberts J. 1982. Limit pricing and entry under incomplete information: An equilibrium analysis. Econometrica 50(2): 443-459. Mollick E. 2012. People and process, suits and innovators: individuals and firm performance. Strategic Management Journal(Forthcoming). Nahm J. 2008. The effects of one-way compatibility on technology adoption in systems markets. Information Economics and Policy 20(3): 269-278. Nair H, Chintagunta P, Dube J. 2004. Empirical analysis of indirect network effects in the market for personal digital assistants. Quantitative Marketing and Economics 2(1): 23-58. Rothaermel FT, Boeker W. 2008. Old technology meets new technology: complementarities, similarities, and alliance formation. Strategic Management Journal 29(1): 47-77. Salop S. 1979. Strategic entry deterrence. American Economic Review 69(2): 335-338. Schilling M. 2002. Technology success and failure in winner-take-all markets: The impact of learning orientation, timing, and network externalities. Academy of Management Journal 45(2): 387-398. Schmalensee R. 1978. Entry deterrence in the ready-to-eat breakfast cereal industry. Bell Journal of Economics 9(2): 305-327. Schumpeter J. 1942. Capitalism, socialism, and democracy. Harper: New York. Shankar V, Bayus B. 2003. Network effects and competition: An empirical analysis of the home video game industry. Strategic Management Journal 24(4): 375-384. Shapiro C. 1996. Antitrust in Network Industries, Address Before the American Law Institute and American Bar Association, 25 January 1996. http://www.justice.gov/atr/public/speeches/0593.htm [24 January 2012]. Shapiro C, Varian H. 1999. Information rules: a strategic guide to the network economy. Harvard Business School Press. Shy O. 1996. Technology revolutions in the presence of network externalities. International Journal of Industrial Organization 14(6): 785-800.

33

Smiley R. 1988. Empirical evidence on strategic entry deterrence. International Journal of Industrial Organization 6(2): 167-180. Srinivasan K. 1991. Multiple market entry, cost signalling and entry deterrence. Management Science 37(12): 1539-1555. Stremersch S, Binken J. 2009. The effect of superstar software on hardware sales in system markets. Journal of Marketing 73(2): 88-104. Suarez FF, Utterback JM. 1995. Dominant designs and the survival of firms. Strategic Management Journal 16(6): 415-430. Sutton J. 1998. Technology and market structure: theory and history. MIT Press: Cambridge. Tripsas M. 1997. Unraveling the process of creative destruction: Complementary assets and incumbent survival in the typesetter industry. Strategic Management Journal 18(S1): 119142. Wade J. 1995. Dynamics of organizational communities and technological bandwagons: An empirical investigation of community evolution in the microprocessor market. Strategic Management Journal 16(S1): 111-133. Waldman M. 1993. A new perspective on planned obsolescence. Quarterly Journal of Economics 108(1): 273-283.

34

Direct effect of backward compatibility

Indirect effect of backward compatibility

Installed base of compatible games for parent console

decreases Software variety for new console

increases

decreases

Hardware demand for new console

Figure 1: Dual effect of backward compatibility

1

GAME BOY POCKET

GAME BOY ADVANCE GAME BOY ADVANCE SP

.8

GAME BOY DS LITE

.6

GAME BOY COLOR

.4

GAME GEAR PSP

.2

DS

0

Market Shares

FIGURES AND TABLES

1995m1

1997m1

1999m1

2001m1

2003m1

2005m1

2007m1

Date GAME BOY GAME GEAR Figure 2: Monthly market shares from 1995 to 2007 GAME BOY POCKET GAME BOY ADVANCE GAME BOY ADVANCE SP GAME BOY COLOR GAME.COM GAME.COM POCKET PRO LYNX 2 N-GAGE N-GAGE QD NEO GEO POCKET COLOR NINTENDO DS NINTENDO DS LITE PLAYSTATION PORTABLE

35

Table 1: Mobile handheld consoles sold between 1995 and 2007 Console

Generation IV Game Boy Game Boy Pocket Game Gear Generation V Game Boy Color Virtual Boy game.com game.com Pocket Pro Generation VI Game Boy Advance Game Boy Advance SP Neo Geo Pocket Color N-Gage N-Gage QD Generation VII DS DS Lite Playstation Portable Playstation Portable Slim

Platform

Game Boy Game Gear

Backward Compatibility

No No

GB Color Virtual Boy

Yes No

game.com

No

Game Boy Advance NGP Color

Yes No

N-Gage

No

DS

Yes

Playstation Portable

No

36

U.S. launch

8/1989 9/1996 1/1991 11/1998 8/1995 9/1997 12/1999 6/2001 3/2003 8/1999 10/2003 8/2004 11/2004 6/2006 3/2005 9/2007

Manufacturer

Hardware CPU Weight [MHz] [g]

Nintendo

4.2

Sega

3.6

Nintendo Nintendo

8.4 20

Tiger

10

Nintendo

16.7

SNK

6.14

Nokia

104

Nintendo

67

Sony

333

300 148 500 188 760 380 n/a 180 142 145 137 143 275 218 280 189

Table 2: Variable definitions Variable

( )

Definition Market share of console j at time t (relative to market potential) Market share of the outside good (no console purchase) Within-group market share (share within the handheld market) Available software titles for current format Deflated console price (1995 prices) Normalized weight of the console Normalized CPU speed of the console Installed base of consoles for the current platform format (millions) Installed base of games for the compatible parent platform (millions) Age of the console (months) Percentage improvement of CPU to compatible parent platform

Table 3: Summary statistics Variable ( (

) ( )

N

Mean

502

) 502

SD

Min

Max

-8.69

2.09 -13.07

-4.69

-2.33

1.97

502 259.04 233.27

-7.71

0.00

3.00 844.00

502

95.94

54.69

502

0.00

0.99

-1.57

3.52

502

-0.01

0.98

-1.44

3.25

502

11.68

10.35

0.00

34.18

502

22.78

31.56

0.00 106.83

502

35.18

30.67

0.00 131.00

502

0.65

0.89

37

20.39 298.23

0.00

3.02

Table 4: Hardware demand estimates (2SLS) DEPENDENT VARIABLE: INDEPENDENT (4-1) VARIABLES SW installed base

( )

( ) (4-2)

0.0149*** (0.00125)

[millions]

Interaction term HW improvement Interaction term Console age Number of available games Deflated price ln(within-group share

( )

Normalized console weight Normalized CPU speed Observations R-squared Hansen’s J

)

-0.0239*** (0.00432) 0.00458*** (0.000608) -0.00885* (0.00454) 0.795*** (0.116) 0.261* (0.155) 0.174 (0.114) 502 0.856 8.171

-0.0109*** (0.00235) 0.000775** (0.000301) -0.00897*** (0.00256) 0.737*** (0.0617) -0.254*** (0.0921) 0.157** (0.0682) 502 0.952 34.37

(4-3)

0.0141** (0.00708) -0.00388*** (0.00139) 0.897*** (0.197) -0.000314*** (9.56e-05) -0.00489** (0.00233) 0.00117** (0.000534) -0.00708** (0.00356) 0.655*** (0.0931) -0.322*** (0.102) 0.169* (0.0854) 502 0.958 34.45

Notes: Standard errors in parentheses are robust to arbitrary heteroskedasticity and arbitrary autocorrelation. Asterisks denote significance levels (*** p